Objective To explore the efficacy of 2 frailty assessment methods, frailty phenotype and frailty index (FI) in assessment of frailty in the elderly so as to provide reference for clinical practice and scientific researches. Methods Totally 106 elderly individuals [at the age of(79.5±7.6)years] who took physical examination in our outpatient department of geriatrics in 2015, were enrolled in the study. After general examination, all the subjects underwent comprehensive geriatric assessment. FI value was calculated and frailty phenotype were assessed for each subject. The prevalence of frailty in the same population was measured by the above approaches, and the relevance or consistency were analyzed based on the results. Receiver operating characteristic (ROC) curves were used to evaluate the effects of FI cut-off points on frailty screening. Results The mean value of FI was 0.19±0.07 in this cohort. According to the results of frailty phenotype assessment, 65 cases (61.3%) were categorized as pre-frailty, 15 cases (14.2%) as frailty, and 26 cases (24.5%) as non-frailty. Both assessment methods indicated that the severity of frailty was increased with age. FI value was positively correlated with frailty level staged by phenotype assessment (r=0.433, P=0.000). When FI cut-off values ranged from 0.09 to 0.25, its consistency (Kappa value) with frailty phenotype was 0.143 (P=0.029), and the area under the curve (AUC) was 0.760 (95%CI:0.616-0.905, P=0.001). While the Kappa value was 0.178 (P=0.002) and the AUC was 0.774 (95%CI:0.629-0.919, P=0.001) when the cut-off value of FI was 0.20-0.35. The optimal cut-off point of FI was 0.19-0.27. Conclusion The elderly at pre-frailty accounts for a larger proportion in this study, and the frailty severity is increased with age. Moderate positive correlation is found between FI value and frailty level by phenotype. The 2 pairs of FI cut-off values can be used to screen frailty for Chinese elderly, but the accuracy is not quite satisfactory.